Following allogeneic blood and marrow transplantation (BMT), mature donor T cells can enhance engraftment, counteract opportunistic infections, and mount graft-versus-tumor (GVT) responses, but at the risk of developing graft-versus-host disease (GVHD). The key to the future of allogeneic BMT as immunotherapy for the treatment of cancer lies in the ability to enhance the beneficial effects of the donor T cells in mediating GVT while minimizing their capacity to cause GVHD. One approach to accomplish this goal would be to selectively deplete subsets of alloreactive T cells in the hematopoietic stem cell inoculum. In this regard, TCR V repertoire analysis by CDR3-size spectratyping can be a powerful tool for the characterization of alloreactive T cell responses. The general aim of this proposal is the investigation of the immunobiology of lethal GVHD and GVT effects by concentrating on how the donor anti-host minor histocompatibility antigen (miHA) T cell repertoire develops and the broad definition of the TCR specificities involved in disease pathogenesis of the target tissues and GVT responses. To accomplish this, the B10.BR->CBA and the C57BL/6 (B6)->CXB-2, B6->BALB.B, B6->CXB-3 and B6->CXB-7 miHA-mismatched strain combination transplantation models will be utilized. Varying levels of GVHD pathogenesis are mediated by CD8+ and/or CD4+ T cells in each of these transplantation models. In addition, the role of host elements outside the hematopoietic compartment, as immune modulators for GVHD will be examined. The T cell repertoires involved in GVHD and GVT responses will be analyzed by TCR V spectratyping. These latter studies will be accomplished by challenging the CBA or CXB-2 mice with host-derived murine myeloid leukemia cells (MMC6 and MME4). The spectratype analysis will be used to guide manipulation of donor T cell inocula in an effort to diminish GVHD and boost the GVT activity. These preclinical murine studies will establish the foundation for the human studies in which TCR V spectratype analysis will be used to examine the predictive value for the in vivo GVT and GVHD responses from patients by analysis of in vitro mixed lymphocyte culture responses, generated between donor T cells and host tumor samples, as well as alloreactivity between donor and host peripheral blood samples (PBL). These in vitro analyses will be compared with TCR V spectratyping of PBL obtained from the patient post-transplantation. Comparing V spectratypes between the in-vitro tumor- and host- reactive responses with those of the in vivo analysis will allow determination of which T cell specificities are most likely to be GVH or GVT reactive-specific.